Beginning in the Fall of 1983, The Bureau of the Census launched a new data collection activity known as the Survey of Income and Program Participation (SIPP). This survey presents researchers with a new opportunity to study issues related to welfare dependency and use. SIPP was designed to be a multipurpose survey providing: 1) improved cross-sectional estimates of annual income, 2) new information concerning assets, liabilities, work and marital histories, child care, taxes, etc., and 3) longitudinal data on changes in labor force activity, income sources, household composition, etc. As such, SIPP's structure differs greatly from the survey most often used to examine the dynamics of welfare, the Panel study of Income Dynamics (PSID). In SIPP, data collection focuses on reporting of income, program participation, and household composition on a monthly accounting period over a relatively short period of time, 2 ½ to 3 years. To reduce recall bias, interviews are conducted at 4-month intervals and questions refer to  each month of the 4-month period preceding the interview month. The PSID obtains data by conducting one interview annually and uses the calendar year as the reference period but has followed individuals over long periods of time with the initial interviews taking place in 1968.

Upon completion of the first full panel file from SIPP, we are now in a position to explore uses of SIPP’s monthly accounting period in studies of welfare dynamics. The purpose of this paper is to present, in a very descriptive fashion, some basic kinds of statistics on welfare dependency so that researchers familiar with this area of work can get a feel for this new data set.

Most of the data compiled for this paper are similar in concept to those previously derived by Duncah (1984) and Bane and Elwood (1985) in their research of welfare issues based on the Panel Study, Obviously, SIPP is not suited to the kinds of analyses these researchers have done using the PSID since the period of time over which persons are interviewed is relatively short for SIPP. The role SIPP plays in the examination of welfare dynamics will, therefore, be much different, and emerge as the data are disseminated and attempts to answer specific questions are undertaken.